In this age of the customer, our clients are increasing their focus on use cases which combine the areas of analytics and personalization to understand customer sentiment. Since Celebrus users capture all customer interaction data in highly granular detail, they understand how this rich data can be used to move beyond merely gaining visibility of their satisfaction levels, to truly understanding their feelings. When armed with this level of insight, organizations can seize the perfect moment to upsell, terminate a potential threat in its tracks or turn a risk of customer churn into an opportunity to form a relationship of lifelong loyalty.
The fact is, being able to gauge and respond effectively to customer sentiment is the holy grail for CX professionals. And if I'm being honest, only a handful of our impressive list of enterprise customers are sufficiently mature to be able to do this. The challenge is huge as it requires an organization to fulfill the academic definition of AI – to achieve fully automated human-like reasoning and interpretation at scale. Well, perhaps that's not an academic definition that you'd find in a textbook, but it's certainly how I picture the challenge.
To provide an example of where I see organizations falling short of this goal, I recently spent some time using a popular messaging mobile application and was having a conversation with a friend on the subject of hay fever. The next day when I logged in to the photo sharing application which is part of the same parent company, I was targeted with a number of ads for a well-known hay fever remedy. Under normal circumstances, this personalization campaign could be considered to be reasonably effective, except for the fact that my earlier conversation had involved me telling my friend that I was relieved and slightly surprised that I had not suffered from hay fever at all this summer, no doubt due to the amount of rain we were having!
This attempt to respond to my specific needs was actually just a crude form of personalization which relies on segmentation based on simple keyword analysis…which in this case led to inappropriate actions. So, was the poor targeting the fault of a simplistic algorithm? Yes, but simply buying a superior decisioning solution would not be enough to solve the problem! To be able to measure and respond instantly to customer sentiment, a best-of-breed decisioning solution (such as Pega, Teradata or SAS) is just one of the 3 capabilities that are required.
The fact is, today's best of breed Artificial Intelligence and decisioning solutions can yield amazingly sophisticated and accurate results when it comes to analyzing customer sentiment. However, despite all of the complex algorithms that they apply, these solution vendors' ability to decipher the meaning behind customer behavior is dependent on the quality of the interaction data that they have to work on. Nothing sums this up better than the much-overused expression: garbage in, garbage out. Any organization who is serious about understanding customer sentiment and turning this insight into a competitive advantage is going to need to commit to capturing the best quality first party experience data possible. Generally speaking, best-of-breed decisioning solutions specialize in generating next-best-actions and therefore do not capture interaction data, which means that organizations will need a best-of-breed data capture and enrichment solution to optimize the benefit they derive from their decisioning solution. If your data capture solution is unable to capture customer interactions across all digital channels (e.g. from mobile apps or IOT devices as well as websites), cannot capture data in sufficient detail or is unable to enrich that data to provide context, then you are at risk of undermining the investment that you made in your decisioning solution.
The third and final capability that is required to effectively leverage insights into customer sentiment, is a genuine real-time capability. The term real-time has become diluted by the number of vendors who make illegitimate claims to the term. Many of these solutions take several minutes, or in some cases, hours to deliver data to its intended destination. While this would not be a problem for more rudimentary analytics or reporting use cases, when using sentiment related insights to maximize opportunities and mitigate risk, the ability to respond in-the-moment is essential. This is simply because organizations have an extremely limited window within which to respond to their customers who are exhibiting signals of opportunity or threat. In these situations, messaging and offers need to be customized to the specific circumstances with a next-best-action which has been generated in response to the underlying meaning behind the customer's behavior. To achieve this, the customer data needs to be available within milliseconds to enable the generation of bespoke content to occur in less time than it takes to load a page within whatever digital channel the customer is engaging with. This is a capability (which very few solutions are able to deliver) is what we call Instant Data.
To find out more about how our customers are using Celebrus and our partner technologies to leverage customer sentiment insights derived from their interactions, contact a member of our knowledgeable team.
